In Our Time - The Infant Brain

Episode Date: March 4, 2010

Melvyn Bragg and guests Usha Goswami, Annette Karmiloff-Smith and Denis Mareschal discuss what new research reveals about the infant brain.For obvious reasons, what happens in the minds of very young, pre-verbal children is elusive. But over the last century, the psychology of early childhood has become a major subject of study. Some scientists and researchers have argued that children develop skills only gradually, others that many of our mental attributes are innate. Sigmund Freud concluded that infants didn't differentiate themselves from their environment. The pioneering Swiss child psychologist Jean Piaget thought babies' perception of the world began as a 'blooming, buzzing confusion' of colour, light and sound, before they developed a more sophisticated worldview, first through the senses and later through symbol. More recent scholars such as the leading American theoretical linguist Noam Chomsky have argued that the fundamentals of language are there from birth. Chomsky has famously argued that all humans have an innate, universally applicable grammar.Over the last ten to twenty years, new research has shed fresh light on important aspects of the infant brain which have long been shrouded in mystery or mired in dispute, from the way we start to learn to speak to the earliest understanding that other people have their own minds. With:Usha Goswami, Professor of Education at the University of Cambridge and Director of its Centre for Neuroscience in Education Annette Karmiloff-Smith, Professorial Research Fellow at the Centre for Brain and Cognitive Development at the Department of Psychological Sciences, Birkbeck College, University of LondonDenis Mareschal, Professor of Psychology at the Centre for Brain and Cognitive Development at Birkbeck College, University of London.

This BBC podcast is supported by ads outside the UK. Every Sunday, we talk about the week's tech news on this week in tech. Hi, this is Leo Leport. Inviting you to join me this week with Lisa Schmeiser, Dan Patterson, and Yanko Rekkers. We're going to talk about the new 49 megabyte web page. It's the standard, you know. We'll also talk about Elon Musk. You've got some spleenin to do and the Yassify filter, new from Nvidia.

That's this week on this week in tech. you'll find it at twit.tv or wherever you get your podcasts. Thanks for downloading the In Our Time podcast. For more details about In Our Time and for our terms of use, please go to BBC.co.com.uk forward slash radio 4. I hope you enjoy the program. Hello, everyone's been a baby. Infantile amnesia means that no one remembers what it's like.

The first two years of life are often described as our formative years, but they've remained a closed mind until recently. The secrets of the infant brain aren't just fascinating, their own sake, many think that they shed light on the fundamental nature of human beings, which is why it isn't surprising the subject has often been a highly contested area for scientific debate. Throughout the 20th century, as the study of early childhood took off, opposing positions of each held sway. In recent years, new research techniques and technologies have been pressed into service to shed more light on this universal mystery. To explore what's recently been discovered about the infant brain, I'm joined by Deney Marischar,

Professor of Psychology at the Centre for Brain. and cognitive development at Birkbeck College, University of London. Usha Goswami, Professor of Education at the University of Cambridge and Director of the University's Centre for Neuroscience in Education. And Annette Carmelov-Smith, professorial research fellow at the Centre for Brain and Cognitive Development at the Department of Psychological Sciences at Birkbeck College and University of London. Annette Kamel of Smith, before we discuss recent discoveries,

can you say what you mean by an infant brain? How old are we talking about? Well, you could start even before birth, but certainly during the first 12 months of life, the brain changes enormously, so the infant brain is very, very early in terms of time. So in terms of our listeners from now on, they're talking about the first 12 months.

12 to 24 months. Unless you specify something else. So that's where we are there. Can you begin by outlining the ideas of Jean Piaget, who died in 1980, his ideas that I understand it were extremely influential. Can you tell us about that? Yes, they were very influential.

So prior to Piaget, really people thought that the brain was a blank slate and that children would learn anything that you taught them. And Piaget came along in the mid-50s from the Anglo-Saxon point of view when Flaville translated his book, The Birth of Intelligence in the Child. And Piaget was, well, he was first a wonderful observer of children, much like Darwin. He registered all his children. activities in diaries. And his theory argued that children constructed their own intelligence,

and by extension constructed their own brains. So in their interaction with the world, they changed the way their brains were structured and adapted to the incoming input. You're talking about him being a great observational scientist like Darwin. Can you give us some idea of what he was observing? He was observing first his own children in great detail, and made extraordinary discoveries. So at one point, a ball runs under a sofa that his son, Laurent, was looking for. And the child stops looking. And Paget tries to understand what an earth that can mean,

does do things disappear from the mind when they disappear out of sight. And from that he devised all sorts of experiments which hundreds of people across the world then used to try and understand what we call object permanence. So he was a very fine observer of the child mind. Can you give us one or two more example, just to get the listeners accustomed to what we're talking about. The sort of fine detail we're talking about.

Absolutely. A little older, so in three, four-year-olds, he would take two glasses, one very wide one and one a very narrow one, and pour water from the wide one into the narrow one, and it would go much higher. And he would ask children, is there more in one of them? And young children would think, because it was higher, there was more liquid, despite the fact that nothing had been added or subtracted.

So he studied the gradual change in children's thinking about conservation of liquid, for instance. He did the same with number, mass, and so forth. And when he made his observations known, was this thought of at the time as being a radical departure in studies of, let's keep calling it, the infant brain? Yes, I mean, he wasn't studying the infant brain. At that point he was studying the infant mind.

He didn't have the techniques to study the infant brain. But it was a radical change because he was a constructivist, So he thought about knowledge as developing over time and that the beginnings of knowledge were crucial for us to understand adult knowledge. And that was a very different position from the empiric's views that had preceded it. In the later 20th century, his ideas were challenged. Can you broadly tell us how they were challenged? I think it really came at first from Chomsky, the American linguist,

who argued that you couldn't learn from the environment alone. and there had to be some innate structures, and for him it was a universal grammar, and that that was common across all languages, despite their superficial differences. We are born with a universal grammar hardwired. That's what he was arguing, and this was very different from Paget's point of view

that you gradually constructed your structures in the brain. But nevertheless, Piazre's development was that it wasn't a clean slate. There was a lot of busyness and sorting out going on. Absolutely. As young as four months or even younger, he was discovering. I think Piazre was a sort of halfway halfway, house between nativism, the sort of idea that things were built in to the brain, and the empiricism that suggested that we had just a blank slate and anything could be imprinted

on it. Denise Marischal, can you set out the view that's emerged of the last two decades? Is there a sense in which these two, we haven't really talked much about Chomsky, nativism, the idea of being hardwired from the start, as it were, being ready for the world when you come into the world? Have these two views in some way merged, or is there some consensus come out of this? Well, I think there's been, in the last 10 or 15 years or so, there's been a return to Piagetian ideas. But these ideas augmented by the new technologies that have allowed us to look in more detail at what behaviors the infants and children showed,

as well as what was going on in the brain at that time. The Chomskyan nativist's view is that a lot of the sophisticated competencies that are found in adults and seem to be attributable to infants from a very young age, are present from the very beginning. The sophisticated neuroimaging technique suggests that that's not quite true. What is true is that fragments of those abilities are present from a very young age, and that as the infant interacts with their environment, they piece together those fragments to gradually construct

the sophisticated understanding of, say, number or hidden objects that we see in adults. Can you just tell us how Piaget thought the brain actually worked? And then how Chomsky, let's take these two. rather simplistically, but fairly authentically, as the two poles. How does he think you work in terms of neurons, synapses and so on? So ironically, neither of them thought about the brain. They didn't have the technology to do so.

So from a chomskin perspective, they're interested in the mind, in thought, in ideas, and they believe that you can study the mind independently of understanding the brain, in the same way as one can study the software of a computer, how, say, a program like Microsoft Word works, without needing to understand the circuits underneath it. There's a dissociation between the functional and the structural levels.

And according to Chomsky and the nativists, what's interesting is the functional level. Piaget, on the other hand, very much had his roots in biology. In fact, he trained as a biologist, and he believed that the biological substrate determined the way our minds grew, but he simply didn't have the technological tools to allow him to investigate that

in any way that would allow it to inform his theories. Of course, much water has gone under the bridge since then. Many discoveries have been made, and we understand a lot more about how the brain works. The brain is both very simple and a very complex system. It's very simple in that the building blocks of the brain are incredibly simple. Those building blocks are cells called neurons, very long, elongated cells, a bit like cables that transmit information in terms of electrical pulses. And all those cells do is simply sense, pay attention to what signals are being sent into them.

If the number of signals exceed some threshold, they start firing as well, sending electrical signals as well. That's all they do. Setting electrical signals on to the next cell. Well, they send electrical signals down their cable. They communicate with other neurons through junctions called synapses. Now, they can't actually make the electrical signal jump across the synapses. What they do is secrete some chemicals called neurotransmitters, which bridge this gap and cause a little signal in the receiving neuron. Now, that's a very simple system.

Simple cells, simple firing of electrical signals. Where the complexity comes in is in the number of neurons that we have and the number of connections that are made between those neurons. There are, in fact, about 10 billion neurons in an adult brain, and most of those, if not all of them, are actually in place from birth. Each of those makes about 1,000 connections, on average a thousand connections. So there's about 10,000 billion connections in the brain. Those are not in place from birth.

And what happens during learning and development, especially in the first few years, is the establishment and the modulation of those connections. And it's those connections, the strength of those connections, that determines the particular functions that parts of the brain will make. So all of this, actually, was unknown to Piaget. Current studies of the infant brain have actually tried to take Piagetian ideas, which were based on biological principles and informed them with the latest understanding of how the brain functions. Uchugoswamas, I understand it.

The breakthrough in research into infant brain came with the idea of studying how long infants look at things. Well, that was one of the initial breakthrough. Can you tell us exactly what that meant and why this was such a powerful new research technique? Because it would seem, well, it's fairly obvious. It's linked to what Denise said, which is that the information has to come into the brain.

Most babies don't start moving on their own until they're about nine months, ten months of age. So certainly for the first six months, looking is one of the main ways to get information about the world. And also touching. So things like manipulating objects is very important, which was recognised by Piaget. But babies do spend a lot of time looking around their environments,

and they're also very interested in when things in the environment change. So things happen. And often things happen because of human agents. And you'll find that babies spend a lot of time watching human beings do things. actually any animate things like dogs or cats. Infants will spend a lot of time looking at. And they like novelty, so they like it when new things happen.

And so you can use that looking bias to start trying to understand either basic questions like when do they understand that an object that's out of view has actually ceased to exist. Like the ball going under the surface. Exactly. And it said at the beginning, yeah. Or you can look at whether they spend a lot of time trying to work out the intention of other actors in their environment.

because we do things for reasons. We have mental intentions that are hidden from view, but that you can reconstrue through people's actions to some extent. And so what babies look at has turned out to give us a wealth of information about how they're building this cognitive system, their own brain. So how do we go about, when you say it's new research, what do they do that they weren't doing before? Can you give us some examples?

I think the looking experiments really started off looking at sensory systems, so trying to understand whether babies could say distinguish different colours. or whether they could tell that a spherical object was different from a cube, this kind of thing. But then it was realised quite quickly that you could use the same technique to either look at cognitive concepts, so things like do objects still exist. You could also use looking behaviour to look at language learning, because if you reward babies looking, so you have some novel event that only happens when they, for example, turn their head to look at it,

then you can look at whether they can distinguish one sound from another, for example. So since about 1980, looking time has been one of the sort of fulcrums of developmental psychology. But in the last five, ten years, of course, we've got the brain imaging methods that Deneem mentioned as well. And you can actually measure directly this electrochemical signaling in the brain by putting sensors on the scalp, on the baby's scalp, so babies have no hair. You can get quite a clean signal from the baby. And you can try and look at brain activity in real time. and also eye-tracking devices which can follow where the baby's pupil is moving in a complex environment

and you can really see what the baby's focus of attention is, which tells you quite a lot about what the baby's likely to be learning. Can you be more specific about these new techniques, new technologies, give us some of the names and what they do specifically? Well, a good example of just liking novelty is looking at when babies first become able to recognise their own name, which it turns out by about four months babies, can recognise their own name in a stream of sound of someone speaking. And that was shown, for example, by training babies.

So babies are very good at learning contingencies. So you can train them that if they turn their head to look at something, something exciting will happen, like say, I don't know, a mechanical toy will start activating. And then you can reward them for looking. So you either selectively play them somebody speaking their own name or somebody speaking other children's names that have very similar phonetic patterns.

And you'll find that the babies preferentially turn their head and look to activate their own name. I'm reading in preparation for this program that there are about 600 basic phonetic sounds and all babies quickly become accustomed to these. Actually, it's more that the way the brain is built, it can recognise all these 600 sounds at birth. At birth?

At birth. How do we know that? In fact, this is a sucking technique, was the classic experiment. So babies like sucking. is another thing babies do a lot. So you can get them to suck on a dummy, and when they suck more strongly,

you can reward with something. So, for example, well, you could reward with the mother's voice, and then the baby will learn to suck very strongly to get that. And then if you hear another female voice, the baby will suck less. And so you can use that same technique to contrast, say, the p-sound and the b-sound, pat versus bat,

to see where the babies can distinguish a change, because if they're sucking away to someone, saying p, p, p, p. It's quite dull, and so they start sucking less and less. But then if you change that phonetic sound to b in the beginning of bat, then they'll start sucking more strongly again. That was the initial way that this was shown. But very quickly, through language experience,

babies will focus just on those contrasts that are used by their native language, which in English might be, say, 45 or 50, of these 600 potential contrasts. So within the first year of learning, based on environmental experience of hearing their language around you babies are specialising their brain equipment to just focus on the sounds that matter

for their language. So they're coming down from 600 to about 45? Yes. Do they lose contact with the other 150? Oh, 555 completely. That's still controversial. So, for example, using the brain imaging technique for picking up electrochemical signaling,

you can show that the effect that's been shown through these sucking type experiments, you can also show through the electrical evoke potential, the way the brain fires these electrical signals in response to say, P versus B. And you can show that when it's a contrast that your language doesn't use, the sixth-month-old baby will show very strong electrical signaling to the contrast. The 12-month-old baby will show much less strong signaling.

However, they'll still show a difference. So the question is whether they're then using that cognitively at all. probably they aren't. Do you want to take that on, Annette, is this a real step-jump forward in research because if you have these techniques and these imaging machines as are? Oh yes, I think it is for two reasons.

One is that we can look at the same behaviour across age and see different brain patterns so we know that actually something is changing. The other thing is that sometimes in overt behaviour you don't get, so 10-month-old babies don't show a disfereign. between non-native sounds. But if we look at the brain activity,

we can see that they do. And in fact, I had a student who looked at adults and found that they were completely unaware that they could hear different Hindi sounds, yet their brain was still registering it. So it's giving us a real insight into kind of the layers at which the brain can process things and what comes into consciousness.

I wish you mentioned something, or spoke of something very dramatic there, moving from 600 sounds to 40. specialist sounds between the age of nothing and about 12 months. Have you any idea now through all this stuff how they arrive at that? Is it just what they hear or something else going on? Yes, no, it's what they hear. But as Deney was saying at the beginning,

connections get strengthened over time when they are very useful to the environment the baby's in. So the mother tongue sounds will be constantly strengthened. Those that are there for other language sounds will be more. weakened. And so we know through that, that over time you will lose that possibility. I think this is another good example

of where our understanding of how the brain works has helped explain what happens in terms of infant behavior. As I said earlier on, the brain learns through modulating connections between neurons. Now, it turns out that how those connections are modulated has to do with whether events co-occur

with each other. So what the brain is really doing is detecting covariance correlations in its environment. It's in effect a statistical inference machine. And if things tend to co-occur repeatedly, the connections associated with that will be strengthens. Can you give this a specific example, the example of which we were discussing, of the particular sounds that are in a child's native environment will be reinforced, so they'll tend to co-occur together. And that will give more muscle to the connections. It will give more muscle to the connections that were used, between the neurons used to encode those sounds. But I just want to say that this is a very general phenomenon. It's not just

in language. So this example of perceptual tuning where perhaps the infant and the infant brain is more receptive to a broad range of stimuli and then it tunes into those that the infant encounters is also found in face recognition. Some recent studies have shown that very young babies can

discriminate individuals both within the human species, individual faces, sorry, between the human species, so different people, as well as within other species, say monkeys. So a six-month- old baby familiarised with the face of one particular monkey will recognise a different monkey as being a different individual. Adults don't do that unless they're experts.

Monkeys look very similar to each other. So the babies who had this ability to discriminate the individual monkeys have lost that ability. In the same ways they might have lost the ability to discriminate different sounds. Yes, whereas monkeys will retain that ability. So the point I wanted to come in on is that this language learning device isn't specific to the human infant brain. because animals too can make these distinctions so there's no accident about where these 600 sound distinctions occur there's something to do with the way sounds are produced anyway

because dolphins or porpoises or chinchillas can also make distinctions like p pah but they of course don't develop language you wanted to go in now yes I also wanted to say another example of this fine tuning is that the baby will recognise the shape of the mouth and the sounds that they hear and very quickly match them So if I go ooh or e, they will match those sounds. And if they mismatch, the baby will look very surprised.

And that's the use of those techniques. So if you do an E and say anew, they'll be surprised. Yes, absolutely. So they're learning all the time and putting things together. And it brings back Piaje as well because of his idea about the importance of action, of motor behaviour. Because language isn't just a sound, it's also an action. And so at the same time, babies are learning to make these sounds themselves. And that's very important as well for the development of this language.

The action means on the action of the mouse. Yeah, and the jaw and the way you articulate. And the hands, about to look at it. Well, gesture is actually your same point, yeah. Well, you're all just. One of the key points of piages is that babies are actually actively engaged in constructing their understanding of the world. So the infant is not just a couch potato. It's not a sponge soaking in information.

The brain picks up correlations, so one might think that all you have to do is watch the world, and your brain will pick out the important things. The problem is there are so many correlations in the world. the brain or the individual has to decide which are the important ones. And it's by acting on the world and observing the consequences of that action that the babies are able to focus in on what are the relevant or most important correlations that occur in the environment. You mentioned at the very beginning or near the beginning,

and the ball rolling up the sofa and had it disappeared or was it still there. And there's a term called object permanence, which played a big part in what you're discussing. Could you develop that, please? the idea of object permanence, when it came in and what it means and what its significance is? The significance is that when anything is in the world and changes, it doesn't have to necessarily disappear, we learn to recognise that it's the same thing.

So you in different angles, for instance. But the important thing that Piazsche was trying to understand was the child's really deep understanding of the physical world, when objects disappeared behind an occluder, did the child understand that they were still there despite their change, when an object changed shape because you'd rolled it out a piece of plasticine. Was it the same as the original one?

So he was trying to get at something very fundamental about what is for an adult, something incredibly obvious. And how did he do that then? And at what age was he succeeding? And this was obviously new stuff when he was doing it. Yes, well, it was during early infancy in the first year of life. He looked very carefully at how children really understood that.

And he found it was quite late. The recent research... Quite late being 12 months. Well, no, even later than that, two years, yes. Whereas the recent research... Quite later what? Sorry, I'm sorry to be so picky, but this is really important. Quite late that they actually got it.

Yes. That a ball that had gone under a surfer was still there. Yes, in the room. In the second year of life, it took me. Sorry, if I can just say, I mean, basically for Piaget, there was a gradual construction of the understanding of object permanence over the first two years of life. So in some situations early on, even very young babies might,

remember that a hidden object continued to exist. So by about seven or eight months of age, if you hid a favorite toy under a pillow, the baby would remove the pillow looking for their favorite toy. So this is evidence that they remembered the toy was still there.

But if you surreptitiously, if you appeared to hide it under the pillow, but serptitiously took it away, when you're eight-month-old, remove the pillow and didn't find it, they acted as if the object no longer existed. They wouldn't continue to search knowing that it must be somewhere,

even if it wasn't under that pillow. So Piaget argued they didn't have a full understanding of object permanence in the same way as an adult has a full understanding. If you take my wallet and hide it and I can't find it anywhere, I still continue to understand that it is somewhere in this world. And it's not until about two that Piaget said babies. If I could just follow that up with some neuro evidence. An example of where the new techniques have shed some light on this comes from some work out of our center, where people recorded, researchers recorded the electrical activity of the infant brains while they watched a number of the surprising disappearance events.

So they showed babies six-month-old's video clips of toy trains going on a track, going into a tunnel, and either coming out at the right time or not coming out, or indeed going into the tunnel, and the tunnel was removed to reveal that the train had completely disappeared. And they found that on occasions where the train had unexpected, disappeared, there were neural markers, electrical markers, basically synchronized their electrical activity in locations in the brain

that corresponded in the adult to locations of processing surprise. Now, the babies weren't showing any overt behavior, but this was suggesting there was nevertheless a neural signature. You're able to image that on the machines. This is going through on the machines. That's correct. That's right. Sparks are flying. Well, it's not quite sparks flying.

No sparks flying. Activities happening. It looks like an eG. Yeah. Yeah, it looks like an eG. So basically, in order to measure these things, the babies will wear a hairnet with a kind of array of sensors

that kind of measures the electrical activity in the brain, but nothing actually comes out for the brain. Women in my part of the world used to wear hair nets every night. Well, it's very similar. And maybe that's where they get all the wisdom from. Usher, one, how did they realize that they, when and how do infants realize? Two big questions.

Yeah. that they have minds and that other people have minds? Again, much earlier than was originally thought. So even someone like Freud thought that babies couldn't distinguish themselves from their environments until about two years of age or so. Because it was thought that babies existed in this completely chaotic world where events would take place, things would go out of view, that meant the things had disappeared and so on.

But in fact, we're realising now that babies very early on, partly because they are so interested in human action and what humans are doing, and they'll spend a lot of time looking at eyes to try and get information from the eyes that babies are sort of figuring out, well, you are some entity that's like me, this is called the like me analogy.

And so therefore, if I'm, for example, reaching for something because I want it, maybe when I see you move your hand to reach for something, it's because you want it. So it's giving some very rudimentary insight into intentions and into actions that have purpose, goal-directed action. Sorry, you're deducing all.

this from the way that your machines are tracking activity on a... This hasn't so far been done in that way, but using the eye tracker, for example. So one of the classic tests of whether you understand that other people have mental activity that you can't see, observe, sort of physically, is called the false belief test. So a good example of a false belief is that if I put my water here and it's hidden, then I go out of the room and you move my water to a new location. When I come back into the room,

I'll think my water should still be where I left it. So I'll remove the hiding place and look for my water in this location. And it was thought that until about the age of four, in fact, children didn't really understand about false belief because that's something that's just existing in the mind of another person. But using these eye tracker machines,

it's been shown that even if you have quite a young baby, so I think these were 12-month-olds, actually, so not so young. but you can look at, you can show them a little vignette, you know, where somebody hides an object they like in location one. They leave the room, someone else moves it to location two, then the person re-enters the room. And even as they're coming back in the room,

the baby's busy doing anticipatory looking at the place where that person thinks the object is and looking at the person. And then if the person, in fact, magically manages to find the object in the place it now is, the baby spends a lot of time looking at that too, because that's wrong. You know, that isn't where that person thought falsely that the object was. Annette, you were going to come in there. Yes, I mean, I think this is so general that we really need to think about it.

If I were to jump to my feet now and rush out of the room, you wouldn't just note I'd gone. All of you would start thinking, has she forgotten to put money in her car, has she left her children standing on the street and so forth, we can't stop ourselves attributing intentionality to others and thinking whatever they do is the function of what they think. think. And what has been studied is this very gradual development of this in children.

How did autism, how did the study of autism play a part in this? Well, a group of researchers in London, having looked at some of the very early work on what we call theory of mind, so this attribution of intention, began to realise that people with autism didn't seem to understand what was going on in their social world. So they started doing these false belief tasks with people with autism and found indeed that they failed and yet children with very low intelligence pass these tests

as if they could work out what was going on in the world but this was a real deficit in people with autism, in sort of attribution of intentions to others so it makes everything incredibly difficult to understand. Imagine watching a film and not being able to attribute

intentionality. But again I think these skills start with these very low level mechanisms like knowing that if someone's looking at something it means that they're interested in it in sort of mental sense following people's gaze. If I point to something, you know, you'd naturally turn your head and look where I'm pointing, but a baby who's at risk for autism won't naturally follow point behaviours.

Babies who are at risk for autism don't spend a lot of time looking at other people's eyes and trying to follow their mental thoughts, as it were, through what they're looking at. I think the idea that these sophisticated skills, such as theory of mind, are built up from more basic building blocks, basic abilities that are easier for young infants to master, such as following eye gaze and so forth, is really a novel way of thinking about the emergence of these complex abilities as compared to, say, the nativist view from 10, 15 years ago

and is in line with what Piaget had initially suggested. And how do they learn the difference between inanimate and animate objects? Oh, that's a really early again capacity. you think about the world, anomatopteryx start off on their own. They can change direction without hitting into anything. Whereas if you roll a ball, somebody has to roll it. It has to hit a wall to change direction, etc. And children are very sensitive to this. So there's been a lovely experiment where a ball is rolling, two balls sort of interact, move towards each other. Then one ball goes along a trajectory around quite a big detour, because there's an obstacle,

in the way and continues. And you show that to babies again and again and again until they get bored. And then you take the obstacle away and the ball either runs straight to the other ball or it goes around the same journey but there's no obstacle to be there. And what babies show is that they're surprised

when they see exactly the same trajectory as before going around a non-existent detour obstacle rather than when it goes straight to it as if they expect that if two things interact and therefore have intentionality, if there's no obstacle in the way, they'll go straight to it. And so babies actually show quite, you know,

I think it's 11 or 12 months, that they understand this very fundamental difference between what inanimate objects can do and what animat or give inanimacy objects can do. Boucher, can we talk about language? As I understand it, Chomsky, Narn Chomsky, came with the idea that there is,

it's hardwired, more or less when we're born, or when we are born, around the world, and that there is, somehow also inside that hardwiring, an international grammar, a basic international grammar. Let's perhaps leave that aside for the moment. In what sense, does that still obtain in any way that languages hardwired there ready?

Yes and no. I think some of the very early brain imaging work with babies where they've tried to see which structures in the brain are showing activity when the baby is listening to language. So, for example, the three-month-old baby. The baby is using the structures that the adult brain will also use to interpret language. one sense there is some structure hardwiring. But on the other hand, you can't build in a device

that will know in advance which sound patterns, so which of those 600 contrasts and which grammatical structures a particular language will have. Instead, it's very much what Denise said earlier. It's all about learning from the environment. So there are many statistical associations in, say, the English language in terms of the kinds of sounds that can comprise words, the order in which those sounds can follow each other. You know, certain sounds don't tend to follow each other, so that will mean that there's a word boundary there because probabilistically, based on the statistics of the sounds that co-occur,

that must be anomalous. And so babies are again using these statistical learning mechanisms to build up a knowledge of their native language, which it will include in that phonological structure, grammatical structure, in the sound-based structure. I'd like to tell you, though, because... I would like to move what you want. Okay.

Fine. Well, it's really interesting that you can't just do that from looking, that this whole notion of intentionality and of sharing meaning with other minds is actually very important for even this phonetic learning of language sounds because you can't learn that from the television. So whereas you can learn some kinds of information from the television, like physical interaction with objects, how to take a part of complex toy, for example, if you do an experiment with babies where,

so there was a nice experiment in America where these American babies, were playing with Chinese-speaking graduate students every day, hearing all these Chinese contrasts that the English language doesn't use. And it was shown that even by the time those babies were 12 months old, they could still hear these Chinese contrasts that we normally have stopped paying attention to by 12 months. But they filmed all this interaction, and then they showed a new group of babies, the same graduate students showing the same toys and interacting in the same way, but through a television screen.

And even though the babies were fascinated, and spent a lot of time touching the screen and attending to the screen, they didn't retain these Mandarin Chinese phonetic contrasts. Can I just ask you, Denise, Piazre thought you learned about the world in a sensory way, and then it was superseded by symbolic learning, and the sensory fell away. Can you address that? Certainly. I mean, a little bit of history, again, going back to Piaje. Piaget really considered himself to be an experimental philosopher.

He was really interested in the questions of epistemology. Where does knowledge come from? Where do we get the sophisticated adult level of knowledge? He believed that one should look at the child developing and implement the principles of biology to understand that. So according to Piaget, the child emerged from being an organism that was entrapped by sensory perception. The newborn baby was really dominated by its reflexes.

If something touched it, it moved, it was a bright sound, it responded to it and so forth. And gradually, it managed to liberate itself from the tyranny of direct perception to reach the kind of adult level of abstract thinking. So we can think about objects through the use of abstract concepts, the kind of platonic table emerges somehow or another. And the debates that we see echoed in development of psychology very much echo the kind of epistemological debates found in philosophy as well, where the nativists will say that all sounds well and good,

but it's actually philosophically impossible to ground these abstract, concepts in sensations. And the current thought, so the current thought is that, well, actually, even adult abstract thought is probably somewhat grounded in sensory perceptions. So new neuroimaging methods looking at what parts of the brain activate when people simply listen to text suggests that the motor part of the brain, the motor cortex, gets activated when you simply listen to words like run or kick. So the participant is lying in a bed, in a scanner, doing nothing, just.

passively listening to words, and yet their motor cortex, the parts responsible for carrying out running and kicking, becomes active when they hear those words. So just simply thinking, visualizing perhaps examples of running and kicking leads to activating those parts of the brains. And it suggests that the conceptual abstract representation is actually at a neural level grounded in those sensory and motor representations. Annette, is there a sense in which these new developments, which are quite radical, just at the beginning of something, you think this is going to get much more sophisticated

and we're going to be able to really get information through neuroscience in this way? I think one thing is we need very many different methods that all address the same question because each method just gives you part of the picture. A second, I really think now we need studies of the changes in the infant brain over longitudinal time.

Most of the studies are cross-sectional. They take babies at different ages. We need to track this really in terms of time, looking at things like the development of white matter tracks, so how different parts of the brain link up together to form networks. We're really at the very beginnings of that. But I think it's going to answer a lot of questions. So what do you see as being possible?

You said all sorts of things. Can you give us a view of it? I think we're going to find out much more about the differences between human and animal intelligence. I think we're still at the very very important. beginnings of that. So looking at our closest cousins, the chimpanzee and ourselves, I think we're going to also understand far more about the type of attention that the baby pays to different types of things in the environment. Again, we have some ideas about that, but we don't have

that in great detail. So now they're inventing a little pram that you can push the baby around with a camera that's going to record exactly what the baby looks at all day long. And we need that kind of very in-depth detail. So far, sorry, Deney, can I just ask for sure. So far in the research, what are the most interesting implications for us as human beings that have come out of adults as well, that have come out of these studies?

I think what's really fascinating, and which is different from what constructive is like Piaget thought, was how many of the ways that the infant and the young child's brain functions is actually very similar to adults. So, for example, it used to be thought that children didn't reason logically. not very young children, not toddlers, not four-year-olds, but later.

But now it's realised that actually children can be just as logical as adults, but they have a lot less knowledge. So sometimes they'll make inferences that are incorrect. And what seems to be so important developmentally is the overall regulation of your system. So it's having enough experience so that you learn all these extra contextual factors that help constrain reasoning. But it's also being able to self-regulate your emotions or being able to function effectively in the face of conflicts

or other sort of hindrances to showing pure logic, that are very important developmentally. So it's not so much the content of the mind is how efficient you can be with your own mind given all the other things that you've got to do as well. And in terms of the brain, again, if you look at the very young infant, the brain is active in both hemispheres, so there's a lot of bilateral activity.

But with time and specialisation of the brain, You see one hemisphere becoming more dominant for language, the other more dominant for number or space. And that takes developmental time to occur. Danny, do you think this over the last 20 years, this sort of research is leading to a radical reappraisal of what the brain is like, not just in infants, what the brain is like from then onwards?

Yes, in the sense that the simple building blocks were understood almost 100 years ago. But what we are reappraising is the real complexity of the vast networks that are present in the brain. And how just from a few very simple building blocks we can get the range of complex behaviors that we observe in adults and stuff. I just want to follow up on a point that Annette made earlier on. One of the really interesting things is that there doesn't seem to be a magic bullet. there isn't some aspect of the human brain which is present only in humans

and not in other animals. We don't have the biggest brains, we don't have the largest cortex, we don't have the brain of the best metabolism and so forth. If you look at individual species, there's always someone else outdoing us in some way. And yet we're the only ones that write Shakespeare plays

and build airplanes. So there must be something else. And that's something else is a complex interaction between our neural abilities. Another unique or dimension of human is that we actually have very extensive developmental periods. In contrast to precocial animals such as gazelles who can walk within hours of minutes of being born, humans have a very extended period, sometimes all the way up into adolescence, depending on mobility, where their brains are tuning to the environment.

And this may well be what is giving us the advantage. Well, thank you very much. Kamelov Smith, Ushu Goswamy and Denis Marachal. I enjoyed that very much. Thank you. And next week we'll be talking about the first century queen of the icenai, Budika, as she's now known. Thanks for listening. If you've enjoyed this Radio 4 podcast,

why not try others, such as Thinking Aloud, where Laurie Taylor discusses the latest social science research. To find out more, visit bbc.co.ukh, forward slash Radio 4.